Search

While soft robots have many attractive features compared to their hard counterparts, developing tractable models for these highly deformable, nonlinear, systems is challenging. In a recent paper, the authors published a non-classic, five-parameter constitutive relation for a rod-based model of a widely used, pneumatically actuated soft robot arm. It is natural to ask if the complexity of the relation can be eliminated by redesigning the actuator? To this end, finite element models and experimental results are used to further explore the five-parameter constitutive relation. For multiple designs of the pneumatically actuated soft robot arm, we are able to demonstrate how finite element models can be employed in place of experiments to specify the constitutive relations and how the relations are scalable by actuator length and applied pressure. Our primary result is the finding that the five-parameter constitutive relation is germane to pneumatically actuated soft robot arms and the parameters for this relation can be determined by three finite element simulations.

The recent surge of interest in soft robotics has led to interesting designs and fabrication of flexible actuators composed of soft matter. Modeling these actuators to obtain quantitative estimates of their dynamics is challenging. In the present paper, a rod-based model for a popular pneumatically activated soft robot arm is developed. The model is based on Euler's theory of the elastica and is arguably the simplest possible model. Through a synthesis of experiment and theory, we find that the constitutive relations needed to accurately capture the deformation of the arm differ considerably from the simple classical relation that the bending moment is linearly proportional to a change in curvature. The present paper also provides a framework to evaluate whether future soft robot actuator designs can be captured using simple models.

Technical applications often include tribological systems in which friction and wear occur. Hard particles, such as sand, are often present in the contact between the surfaces of two bodies in contact and affect both the system dynamics and the resulting surface topography of the contacting bodies. Real tribological systems can contain a very high number of particles. Hence, for performance reasons of numerical studies, the explicit representation of each particle in a simulation is often not practical. The approach presented here is based on a model simplification by grouping several particles to form statistical sample particles (SSPs). SSPs can only move as a whole, and thereby simplify the motions of the individual particles, but the contact stiffness and orientation dependence of the force are still determined by the set of individual particles that are part of the statistical sample. Using sample size 1, each individual particle is represented without simplification, while using a sample size larger than one, the total number of particles in the system reduces to a minimum of one SSP representing all particles in the system. This approach provides a very efficient way to specify the model complexity and desired level of detail for the particle representation within a model of a tribological system. [ABSTRACT FROM AUTHOR]

Fluids in narrow gaps are employed frequently in many applications. The motivation for their use is diverse and ranges from hydrodynamic lubrication in plain bearings to the transport of hard particles into the working gap for the purpose of machining workpiece surfaces in lapping processes. Depending on the focus of the analysis, it may be useful to investigate the entire pressure field or to calculate only individual quantities. For example, in sophisticated simulations it may be of interest to know the resulting force of a fluid as a function of the external system state in order to describe its damping characteristics. Especially for the simulation of flows in narrow gaps, the Reynolds equation is a convenient choice, which, in contrast to the more general Navier‐Stokes equations, can lead to considerable savings in computational time because no three‐dimensional discretization is required, but only a two‐dimensional discretization. However, if not a highly detailed pressure field is of interest, but only simple relations such as the resulting force as a function of distance and velocity, and if this relation to be evaluated many times for different parameter combinations over a wide range of values, the use of a robust substitute model is a good choice. This article deals with the creation of such a substitute model based on the Reynolds equation taking cavitation into account. [ABSTRACT FROM AUTHOR]

In selective laser melting (SLM), a powdered material is locally melted by a laser and, after cooling, forms a coherent solid structure that enables the production of complex geometries with various materials. The process involves extreme heating and cooling rates and, thus, large temperature gradients, which lead to anisotropic material properties on the macroscopic scale and, in the worst case, reduced mechanical properties. In order to reliably predict the final mechanical component properties, simulations can be performed at different time and length scales. Enormous computational resources are often required to perform such simulations. In order to transform these simulations into suitable surrogate models, the generated data must be compressed and evaluated in a suitable way. This paper shows first preliminary work and a possible new data description of such simulations. [ABSTRACT FROM AUTHOR]